Alkylthio-functionalized polymers containing alternating phenylene and thiophenediyl/bithiophenediyl repeating units: The synthesis and characterization

Article abstract of DOI:10.1246/bcsj.76.1449

Alkylthio(SR)-functionalized polymers poly (3-alkylthio-2,5-thiophenediyl-1,4-phenylene-alt-4-alkylthio-2, 5-thiophenediyl-1,4-phenylene) (P1, alkyl = butyl, octyl and dodecyl) and poly [4,4′-bis(alkylthio)-2,2′-bithiophene-5,5′-diylalt-1, 4-phenylene] (P2, alkyl = butyl, octyl and dodecyl) have been synthesized by Grignard coupling and FeCl₃ oxidation methods, respectively. Spectral measurements indicate that they are highly fluorescent, with quantum yields comparable to those of their alkyl(R)-functionalized analogues, poly(3-alkyl-2,5-thiophenediyl-1,4-phenylene-alt-4-alkyl-2,5-thiophenediyl-1, 4-phenylene) (P1′, alkyl = butyl, octyl and dodecyl) and poly(4,4′-dialkyl-2,2′-bithiophene-5,5′-diyl-alt-1, 4-phenylene) (P2′, alkyl = butyl, octyl and dodecyl). P1 and P2 are thermally less stable than P1′ and P2′, because of the alkylthio group instead of an alkyl group. Modulated differential scanning calorimetry (MDSC) studies reveal a plasticizing effect exerted by the long side chains on the glass transitions (T_g) of the polymers. The P2 series of polymers exhibit a much higher T_g than the P1 series, ascribable to a greater resistance to segmental rotation in the former structure. A cyclic voltammetric study of polymers indicates that they are n-dopable. When doped by I₂ or FeCl₃, high doping levels of the polymers can be deduced from the weight change, XPS and elemental analysis, but conductivities are not high.

Full text of DOI:10.1246/bcsj.76.1449

ꢀ 2003 The Chemical Society of Japan
Bull. Chem. Soc. Jpn., 76, 1449–1458 (2003) 1449
Alkylthio-functionalized Polymers Containing Alternating Phenylene
and Thiophenediyl/Bithiophenediyl Repeating Units: the Synthesis and
Characterization
ꢀ
Jingmei Xu, Siu Choon Ng, and Hardy Sze On Chan
Department of Chemistry, National University of Singapore, Singapore 119260
(Received July 8, 2002)
Alkylthio(SR)-functionalized polymers poly(3-alkylthio-2,5-thiophenediyl-1,4-phenylene-alt-4-alkylthio-2,5-thio-
phenediyl-1,4-phenylene) (P1, alkyl = butyl, octyl and dodecyl) and poly[4,40-bis(alkylthio)-2,20-bithiophene-5,50-diyl-
alt-1,4-phenylene] (P2, alkyl = butyl, octyl and dodecyl) have been synthesized by Grignard coupling and FeCl3
oxidation methods, respectively. Spectral measurements indicate that they are highly fluorescent, with quantum yields
comparable to those of their alkyl(R)-functionalized analogues, poly(3-alkyl-2,5-thiophenediyl-1,4-phenylene-alt-4-al-
kyl-2,5-thiophenediyl-1,4-phenylene) (P10, alkyl = butyl, octyl and dodecyl) and poly(4,40-dialkyl-2,20-bithiophene-
5,50-diyl-alt-1,4-phenylene) (P20, alkyl = butyl, octyl and dodecyl). P1 and P2 are thermally less stable than P10 and
P20, because of the alkylthio group instead of an alkyl group. Modulated differential scanning calorimetry (MDSC)
studies reveal a plasticizing effect exerted by the long side chains on the glass transitions (Tg) of the polymers. The P2
series of polymers exhibit a much higher Tg than the P1 series, ascribable to a greater resistance to segmental rotation in
the former structure. A cyclic voltammetric study of polymers indicates that they are n-dopable. When doped by I2 or
FeCl3, high doping levels of the polymers can be deduced from the weight change, XPS and elemental analysis, but
conductivities are not high.
Conjugated polymers exhibit interesting electrical, electro-
chemical and optical properties because of their spatially ex-
tended p system. However, unsubstituted conjugated poly-
mers are insoluble and infusible. This is a significant dis-
advantage in their characterization and plausible technolog-
ical applications. A breakthrough was made by incorporating
long flexible side chains to the conjugated backbone. Poly-
thiophene and its 3-substituted derivatives represent one fam-
ily of this kind of polymer, in which the substituents can
be alkyl(R),1;2 alkoxy(OR),3;4 alkylthio(SR),5;6 alkanesulfonic
acid (RSO3H)7 or aromatic8 groups. Among them, the elec-
tron-donating SR-substituted polythiophenes have attracted a
lot of attention5;6;9{11 because the electron-donating groups are
expected to produce polythiophene derivatives with low band
gaps through raising the energy level of the highest occupied
orbital (HOMO). This decreases the oxidation potentials of
these polymers, and consequently stabilizes their conducting
states.9
Usually, a high degree of p-electron conjugation and a co-
planar polymer backbone favor high conductivity when
doped.12 However, substituents like SR or R groups on the
thiophene rings tend to cause a steric hindrance and, conse-
quently, a lower degree of coplanarity along the polymer back-
bone as compared to the unsubstituted parent polymers.12
Amongst the SR and R functional groups, because SR is more
flexible than R, SR-functionalized polymers are expected to
have higher coplanarity than their R-functionalized analogues,
and thus a higher conductivity.
decay of singlet excitons to the ground states of the polymers
leads to light emission. The polymers able to realize these
processes are fluorescent polymers. There are several ways to
increase the electroluminescence efficiency through polymer
structure modification, including incorporating non-conjugated
fractions along the polymer backbone so that excitons are con-
fined and the non-radiative decay of singlet excitons is
suppressed;13 modulating the positions of the band edges of
the polymer through derivatization of the polymer backbone,
for instance with electron-donating or -accepting groups, so
that the energy barrier between the Fermi energy of the cath-
ode (anode) and the energy of the frontier levels of the emis-
sive polymers are reduced, and thus an efficient charge injec-
tion can be achieved;13 increase of the torsional angle between
the adjacent units so that a polymer backbone structure with
shorter mean conjugation length is formed and relaxation from
the excited state through non-radiative (e.g. thermal) processes
is reduced.14;15 Based on the above summary, SR-function-
alized conjugated polymers are expected to have a lower band-
gap energy, a higher conductivity and a lower fluorescence
quantum yield as compared with their R-functionalized ana-
logues.
During studies of SR-functionalized polythiophene deriva-
tives, although the fluorescent properties of this kind of con-
ducting polymers had been noted and reported,6;11 only one
report gave information about the quantum yield of this kind of
polymer.6 Our research group has ongoing interest in studying
the structure-property correlation of conjugated polymers. It is
noted that polymerization of monomers incorporating thiophe-
nediyl and acetylene16;17 has led to polymers showing very
If singlet excitons are formed after semiconducting poly-
mers are excited by photo- or electro-stimulation, the radiative
Published on the web July 15, 2003; DOI 10.1246/bcsj.76.1449

1450 Bull. Chem. Soc. Jpn., 76, No. 7 (2003)
Alkylthio-functionalized Polymers
good fluorescence due to the insertion of acetylene units, in
addition to the attained structural regioregularity. We studied
the synthesis and characterization of the polymers comprising
alternating phenylene and alkyl-substituted thiophenediyl/bi-
thiophenediyl repeating units (P10 and P20, alkyl = butyl, octyl
and dodecyl).18{20 A higher fluorescence quantum yield and a
blue shift in the emission wavelength were observed. We
therefore designed and synthesized two series of polymers of
similar structure incorporating phenylene and thiophenediyl/
bithiophenediyl repeating units, but with alkylthio substituents
replacing the alkyl side chains: P1 and P2. The objective was
to prepare new highly fluorescent polymers functionalized
with SR groups and to further understand the structure-prop-
erty correlation. The structures of polymers P1 and P2 as well
as P10 and P20 are shown in Chart 1. The results indicate that
the SR-functionalized polymers are highly fluorescent with
quantum yields comparable to those of the R-functionalized
analogues. The detailed influence of a structural modification
by replacing R with SR on the polymer properties are reported
and discussed in the following, compared with their R-func-
tionalized analogues.
tively, as shown in Scheme 1. The syntheses of P1 and P2
were more difficult than their R-functionalized analogues, P10
and P20, because of the replacing of R by SR. P2 are brown or
brownish green powders, which are partially soluble in CS2,
CHCl3, THF and toluene. The solubility of P2 polymers de-
creases with an increase in the alkylthio chain length. P1a is a
dark-red powder, but P1b and P1c are dark-red sticky
materials. They can be easily dissolved in the above-men-
tioned solvents. Films of P1 evaporatively cast from toluene
solution are transparent and smooth, but those of P2 are not.
Table 1 summarizes the number-averaged molecular
weights (Mn) of the polymers, as determined by GPC using
polystyrene as a standard. The Mn for P1a is 3700, corre-
sponding to about 30 aromatic units in a polymer chain. Mn
for P1b and P1c are around 2900, indicating a lower degree of
polymerization due to the lower reactivity of the monomers
with an increase in the side chain length. Mn for P2 is in the
range of 2600–3000, contributed mainly by the lower molecu-
lar weight components, because the polymers are only very
slightly soluble in THF. The insoluble part of the polymers is
expected to have a higher molecular weight. The moderate
molecular weights of the polymers are at the same order as
those of the SR-functionalized polythiophenes prepared by
different methods (R = ethylthio, bis(ethylthio) and ethylene-
dithio, Mn ¼ 2200, 2600 and 3030/4750, respectively).5;6;10;11
Results and Discussion
Physical Properties. P1 and P2 were synthesized by
Grignard coupling and the FeCl3 oxidation methods, respec-
Chart 1. Polymer structure of P1: poly(3-alkylthio-2,5-thiophenediyl-1,4-phenylene-alt-4-alkylthio-2,5-thiophenediyl-1,4-phenyl-
ene) and P2: poly[4,40-bis(alkylthio)-2,20-bithiophene-5,50-diyl-alt-1,4-phenylene].
Scheme 1. Reagent and conditions: (i) 4 molar amounts bromine, AcOH/CHCl3; (ii) zinc dust, 2-propanol; (iii) sodium 1-alkane-
thiolate, CuO, KI, DMSO; (iv) 4 molar amounts FeCl3, CHCl3; (v) 2 molar amounts bromine, AcOH/CHCl3; (vi) (a) Mg, 1,2-
dibromoethane, THF; (b) 1,4-dibromobenzene, Ni(dppp)Cl2, THF.

J. M. Xu et al.
Bull. Chem. Soc. Jpn., 76, No. 7 (2003) 1451
peating unit contains two S atoms in P10 and P20). Residue
bromine and chlorine elements were also detected in polymers
P1 and P2, obtained when the brominated monomers were not
completely converted into Grignard reagents and when chlo-
rination occurred during FeCl3 oxidation, respectively. The
ratio of aromatic rings to Br or Cl atoms was calculated and
compiled in Table 2. However, as compared to their R-func-
tionalized analogues, P1a and P2a do not display a significant
difference in the ratios.
Table 1. Summary of Molecular Weights (Mn), Polydisper-
sity Indices (PDI) and Conductivities (ꢁ) of P1 and P2
GPC results
ꢁ/S cmꢁ1a)
Polymer
Mn
PDI
doped by I2
5:0 ꢂ 10ꢁ4
doped by FeCl3
b)
c)
c)
P1a
P1b
P1c
P2a
P2b
P2c
P1a0
P2a0
3700
2900
3000
2800
2700
2900
3900
6700
2.1
1.5
1.5
1.6
1.7
2.5
2.2
2.7
—
—
—
c)
—
c)
—
1:7 ꢂ 10ꢁ5
1:3 ꢂ 10ꢁ5
2:5 ꢂ 10ꢁ5
7:4 ꢂ 10ꢁ4
2:0 ꢂ 10ꢁ1
1:0 ꢂ 10ꢁ4
b)
—
Structure Characterization. Figure 1 shows the 1H NMR
spectra of monomer 1,4-bis(3-butylthio-2-thienyl)benzene
(4a), polymers P1a and P2a. Because of the structural simi-
1:0 ꢂ 10ꢁ3
1:0 ꢂ 10ꢁ3
1:1 ꢂ 10ꢁ3
1
larity, the overall line shapes of the H NMR spectra of 4a,
a) The insoluble part of the polymers were used in conductiv-
ity experiments as the soluble part is not enough. b) Samples
are too brittle to be measured. c) Neutral polymers are sticky
materials.
P1a and P2a are close to those of their R-functionalized ana-
logues,20 except that the resonances are mostly shifted down-
field due to the replacement of R by SR. Thus, the assignment
of the resonances is not given in detail here. The information
1
from H NMR spectra suggests that polymers P1 and P2 have
The polymers are conductive (the insoluble part of the poly-
mers was used in conductivity experiments because the soluble
part was insufficient) when doped by I2 or FeCl3, as shown in
Table 1, accompanied by a color change from brown to black.
An I2 uptake up to 200 w/w% is possible in some cases.
However, their conductivities are lower than those of the al-
kyl-functionalized polymers. The polymers are expected to be
more conductive because their substituent, SR, is a stronger
electron-donating group than R. The reason for this unexpect-
ed result is discussed later.
the expected 2,3,5-trisubstituted thiophene units and 1,4-di-
substituted benzene rings in their polymer backbone. The
polymers were obtained as expected with regioregular struc-
tures, and no cross-linking occurred during polymerization.
Moderate molecular weights for P1 and P2 are also suggested,
because the two sets of doublets at ꢀ 7.10 and 7.30 can be seen
in the aromatic range, ascribable to the Cb-H and Ca-H reso-
nances on the terminal thiophenediyl rings, respectively. In
the FTIR spectra (not shown), regioregular polymer structures
were also suggested, in agreement with the results from NMR
studies. The existence of sulfur is indicated by strong C–S
Table 2 gives the elemental analysis results for neutral and
FeCl3-doped polymers. It can be seen that the ratio of chlorine
to iron is around 4:1 from microanalyses of FeCl3-doped P1
and Pꢁ2, indicating the insertion of an anion in the form of
stretching at 627 cmꢁ1 5
.
UV-Vis Absorption and Fluorescence Spectra. Just like
the polymers P10 and P20, which are functionalized with alkyl
groups, the polymers P1 and P2 are also highly fluorescent.
They absorb light at around 400 nm, corresponding to their p–
FeCl4
.
This is consistent with the previously reported
results.21{23 The ratios of FeCl4 to S from microanalyses
ꢁ
ꢁ
ꢀ
transitions, but emit green light strongly at about 520 nm,
suggest that one FeCl4 is roughly associated with one repeat-
p
ing unit (one repeating unit contains four S atoms in P1 and
P2) along the polymer chain. These values are higher than
those of the R-functionalized analogues, P10 and P20, in which
one repeating unit corresponds to about 0.6 dopants (one re-
which is discernible to the naked eye. The peak wavelengths,
bandgaps and relative quantum yields with quinine sulfate as a
reference, are summarized in Table 3. Those of P1b0 and P2b0
are included for a comparison. Because P1 polymers can be
Table 2. Elemental Composition Determined from Microanalysis
Expected elemental
composition
Determined elemental
composition
Ratio of aromatic units
to Br or Cl atoms
Polymer
a)
P1a
C7H7S1
C6:01H7:30S1:00Br0:05
6:17H8:00S1:00Fe0:33Cl1:18
20:1
b)
C
C8:13H13:02S1:00Br0:15
a)
P1b
P1c
P2a
C9H11S1
C44H60S4
C5:5H6S1
6.7:1
c)
—
C5:05H5:47S1:00Cl0:02
5:57H4:61S1:00Fe0:26Cl0:96
a)
37.5:1
18.8:1
9.4:1
b)
b)
b)
C
a)
P2b
P2c
C7:5H10S1
C9:5H14S1
C14H14S1
C11H12S1
C6:38H10:65S1:00Cl0:04
6:52H7:00S1:00Fe0:29Cl1:00
C
a)
C7:38H10:50S1:00Cl0:08
7:11H9:16S1:00Fe0:31Cl0:98
C
P1a0
P2a0
C12:60H14:42S1:00Br0:08
13:04H13:80S1:00Fe0:33Cl1:46
25:1
a)
b)
b)
C
C11:70H12:51S1:00Cl0:038
C10:68H13:86S1:00Fe0:28Cl1:15
a)
39.4:1
a) Neutral polymers. b) Powder polymers chemically doped by FeCl3. c) Not enough P1c
to be measured.

1452 Bull. Chem. Soc. Jpn., 76, No. 7 (2003)
Alkylthio-functionalized Polymers
those only in CS2 are reported for the P2 series. The solvent
CS2 seems to favor a more extended polymer conformation
than CHCl3, since both the ꢂmax and ꢂem of P1 polymers are
red-shifted in CS2 with reference to those in CHCl3. The
relative quantum yields of P1 in CS2 are all higher than those
in CHCl3.
It can be seen that replacing R with SR causes a slight red
shift in the absorption and emission peak wavelengths.
A
lower bandgap of SR-substituted polymers than that of the
R-substituted analogues is also observed, as expected.
The fluorescent quantum yields of P1 and P2 in the solution
state are in the range of 23–56% relative to quinine sulfate,
comparable to those of P10 and P20 (19–69%).20 This is in
contrast to the expectation that P1 and P2 should have a higher
coplanarity along the polymer backbone than P10 and P20, and
thus a lower quantum yield, because SR is more flexible than
R. The fluorescence quantum yield of poly(3-butylthio-2,5-
thiophenediyl) in chloroform was reported to be greater than
0.01,6 corresponding to a relative quantum yield of about >
1.8% with reference to quinine sulfate. It is obvious that the
current polymers incorporating phenylene and thiophenediyl/
bithiophenediyl repeating units have higher quantum yields.
This effect can be attributed to the phenylene moieties contri-
buting to a more rigid polymer structure, with the effect that
the relaxation from the excited state through the non-radiative
(e.g. thermal) process would be reduced with consequently
higher fluorescent quantum yields.15;24
Fig. 1. 1H NMR spectra of (a) 4a, (b) P1a, (c) P2a.
Literature reports about the fluorescent characteristic of SR-
functionalized poly(thiophene) derivatives are not as many as
those of R- and OR-functionalized ones. To the best of our
knowledge, only poly(3-butylthio-2,5-thiophenediyl),6 poly-
(3,4-ethylenedithio-2,5-thiophenediyl)11 and the copolymers
incorporating alkylthiothiophenediyl and 4,40-biphenylene re-
peating units were reported to be fluorescent.25 The emission
wavelength maxima (ꢂem) of the former two polymers were
located at 603 and 552 nm, respectively, in the solution
state.6;11 These are red shifted, as compared to those of P1
and P2 (510–520 nm) in solution. On the other hand, ꢂem of
the copolymers incorporating alkylthiothiophenediyl and 4,40-
biphenylene repeating units peaked at 485 nm in the film
state,25 which is blue shifted as compared to those of P1 and
P2 (520–530 nm) films. These results suggest a trend: the
insertion of phenylene units tends to make ꢂem blue shifted
and the more the phenylene units in the polymer backbone, the
shorter the ꢂem of the polymers. The bandgaps of P1 and P2
are in the range of 2.34 to 2.43 eV, in between those of ethyl-
thio-, bis(ethylthio)-, ethylenedithio-substituted polythio-
phenes (2.0 to 2.24 eV)11 and the polymer comprising 4,40-
biphenylene and 3-alkylthio-2,5-thiophenediyl units (2.57
eV).25 This suggests that with the insertion of phenylene moi-
eties, the polymer has a bigger torsional angle or shorter mean
conjugation length along the polymer backbone, and conse-
quently a higher bandgap energy.
Table 3. Summary of Results from Absorption and Emis-
sion Spectra of P1 and P2 Polymers in Solution and Solid
States
Quantum Optical
yield/%a) bandgap/eV
Sample
ꢂmax/nm ꢂem/nm
CHCl3 solution P1a
400
396
396
510
510
510
488
494
25
33
23
66
23
P1b
P1c
P1b0 386
P2b0 398
CS2 solution P1a
416
414
406
408
412
416
522
524
524
525
523
523
30
37
31
44
39
56
P1b
P1c
P2a
P2b
P2c
film
P1a
P1b
P1c
410
390
390
530
526
526
512
524
524
520
521
2.37
2.38
2.34
2.64
2.43
2.40
2.38
2.58
P1b0 388
P2a
P2b
P2c
410
420
426
P2b0 420
Thermogravimetry (TG) and Modulated Differential
Scanning Calorimetry (MDSC). The thermooxidative sta-
bility of functional polymers is an important property to con-
sider in their applications as device materials. The thermal
stability of P1 and P2 was examined by TG, in comparison
with that of the R-functionalized polymers. Just as the R-
a) Relative to quinine sulfate in 0.1 M H2SO4.
dissolved in both CHCl3 and CS2 solvents and P2 have higher
solubility in CS2 than in CHCl3, the optical data obtained in
CHCl3 and CS2 solutions are presented for the P1 series, while

J. M. Xu et al.
Bull. Chem. Soc. Jpn., 76, No. 7 (2003) 1453
Table 4. Glass Transition Temperatures (Tg) of P1, P2 and
Their R-Functionalized Analogues P10 and P20 as Deter-
mined from MDSC
Polymer
Tg/ꢃC
Polymer
Tg/ꢃC
P1a
P1b
P1c
P2a
P2b
P2c
26.4
ꢁ38:8
P1a0
P1b0
P1c0
P2a0
P2b0
P2c0
58.3
ꢁ4:0
ꢁ19:2
122.3
50.4
a)
—
129.0
55.3
65.4
59.6
Fig. 2. TG plots of (a) P2a, (b) P2a0, (c) P1a, (d) P1a0 in air.
a) Polymer is too little to be measured.
substituted polymers reported in our previous work, polymers
with shorter side chains are generally more thermally stable
than those with longer chains. However, the decomposition
behaviors of SR- and R-functionalized polymers are different.
Figure 2 shows the TG thermograms of P2a (a), P2a0 (b), P1a
(c) and P1a0 (d) in air. Two main decomposition steps are
involved. The first is due to the cleavage of a substituent
chain, the second to degradation of the polymer backbone.
In the present series of polymers, the decomposition of the
first step starts at significantly lower temperatures for the
SR-functionalized polymers (thermogram a), as compared to
the corresponding R-substituted ones (thermogram b). A pos-
sible reason is that the thio etheral linkages between the alkyl
group and thiophenediyl rings are weak sites susceptible to
thermooxidative attack. The difference in bond energies of
C–C (317 KJ/mol) and C–S (290 KJ/mol) may serve as a
support to this prediction.26 Interestingly, the order of stability
of P1a (thermogram c) and P1a0 (thermogram d) reversed
above 550 ꢃC. It was reported that sulfur-containing com-
pounds might undergo pyrolysis to produce sulfur-containing
radicals, ArSꢄ, which could dimerize to afford additional com-
pounds, such as ArSSAr, or couple with Arꢄ to form ArSAr.27
These cross-linking reactions can lead to the formation of
pseudo-ladder polymers that may be responsible for the high
thermal stability of P1a over P1a0 at high temperatures. In
addition, the thermal stability of P1a (thermogram c) is better
than that of P2a (thermogram a), in parallel with the superior
thermal stability of P1a0 (thermogram d) to that of P2a0 (ther-
mogram b).
Fig. 3. CV curve of P1a. The scan rate is 50 mV sꢁ1
.
by the pendant groups, which are in closer proximity in the
former series.
Cyclic Voltammetry. The electrochemistry of toluene cast
films of the P1 and P2 polymers was investigated using cyclic
voltammetry (CV) in a 1 M acetonitrile solution of Bu4NBF4.
P1 polymers are all concurrently p- and n-dopable, accompa-
nied by a color change from yellow to black and blue,
respectively. The CV curves at both anodic (p-doping) and
cathodic (n-doping) directions obtained for P1a are shown in
Fig. 3. The limited reversibility of the oxidation and reduction
processes is shown in both anodic and cathodic sweeps, prob-
ably due to degradation of the polymers. Repeated scanning or
scanning to a wider potential range deactivated the film.
Nevertheless, the reversibility and stability were good at mod-
erate oxidation or reduction potentials.
MDSC20 was employed to study the technologically impor-
tant glass-to-rubber transition of the current series of polymers.
Well-defined glass transitions of the polymers were observed,
the results are summarized in Table 4. Tg(s) for the P10 and
P20 series are also included for a comparison. The results
show that in both series of polymers the introduction of a long
side chain (going from butyl, octyl to dodecyl) lowers Tg
significantly. This can be understood because the onset of
the glass transition is associated with the start of rotation of
the C–C bonds along the polymer backbone. The long side
chains push the neighboring polymer chains away, hence re-
ducing the intermolecular attraction, acting as an internal lu-
bricant, or plasticizer. For the P1 series, the SR substituent
significantly lowers Tg compared to the R analogues, but the
opposite is observed for the P2 series, although the difference
is much smaller. The results also clearly indicate that the P20
and P2 series have much higher Tgs than their P10 and P1
analogues. This is due to the greater steric hindrance exerted
For P2 films, the occurrence of n-doping was accompanied
by a very sharp color change from brown to blue or greenish-
blue, although no well-defined onset or peak potential was
observed in their CV profiles. This was because the n-doped
P2 polymers were very easily soluble in acetonitrile. In addi-
tion, a very serious ‘‘break-in’’ phenomenon28 was observed
during cathodic scanning due to the poor quality of the films.
However, no color change was observed in the poorly defined
p-doping process.
Table 5 summarizes the onset and peak potentials for both
the p- and n-doping processes of the P1 polymers. Those of
P1a0 are also included for a comparison. Bandgaps evaluated
as the difference of the onset potentials of the two processes
are also listed. With an increase in the side chain length, both
the p- and n-doping onset potentials increase, resulting in al-
most unchanged bandgaps. In addition, the electrochemical
bandgaps are about 0.3 eV higher than the optical ones
(Table 2), as observed and discussed earlier.20

1454 Bull. Chem. Soc. Jpn., 76, No. 7 (2003)
Alkylthio-functionalized Polymers
Table 5. Summary of CV Results
p-doping
n-doping
Polymer
Eg/eV
Eon/V Epa/V Epc/V Eon/V Epa/V Epc/V
c)
P1a
1.01 1.42 0.84 ꢁ1:59
—
—
ꢁ1:71 2.60
c)
c)
P1ba)
P1ca,b)
P1a0
1.11 1.72
1.23 1.61
—
—
ꢁ1:50
ꢁ1:67 2.61
c)
c)
ꢁ1:34 ꢁ1:78
—
2.57
1.05 1.33 0.60 ꢁ1:87 ꢁ2:14 ꢁ1:88 2.92
a) Degradation occurred in both p- and n-doping processes. b)
Film peeled from electrode surface and some of the readings
may be unreliable. c) Unresolved.
Fig. 5. In situ potential dependent UV-Vis-NIR spectra of
P1a during n-doping process.
potentials were applied, as shown in Fig. 5 for P1a. Two new
broad bands at 460 nm and > 1400 nm, which are different
from those in p-doping (Fig. 4), were observed when reduced
at ꢁ2:2 V accompanied by the color change from brown to
blue. A dramatic spectral change (not shown) was also ob-
served when a cathodic potential (e.g., ꢁ2:2 V) was applied to
P2 polymers, although the doped polymers dissolved in the
solvent easily.
XPS. XPS has been used to study the different electronic
environments experienced by different atoms, and to calculate
the doping level in the present polymers. Figure 6(a) and (b)
show the S(2p) core level spectra of P2b in the neutral and
FeCl3-doped states, respectively. The S(2p) spectrum decon-
volution for the neutral polymer involves two spin-orbit split-
ting doublets [S(2p3=2) and S(2p1=2), ꢀE ¼ 1:2 eV, area ratio
= 2:1], with BE of S(2p3=2) peaks at 163.1 and 164.0 eV,
respectively. The two different environments have been as-
signed to the sulfur atoms on the alkylthio side chain (Sa) and
on the thiophenediyl ring (Sr),29 respectively. When the poly-
mers are doped, four chemical environments are best decon-
voluted from the S(2p) envelope. The S(2p3=2) binding ener-
gies for the four S species are at 163.1, 164.0, 165.1 and 166.7
eV, respectively. While the first two environments are attrib-
uted to the neutral sulfur atoms on the alkylthio group and
thiophenediyl ring, the BE at 165.1 eV is probably associated
with the ring sulfur species, which is positively charged due to
Fig. 4. In situ potential dependent UV-Vis-NIR spectra of
P1a during p-doping process.
In comparison with P1a0, the electrochemical bandgap of
P1a is smaller by 0.32 eV. This is in accordance with the trend
obtained in optical bandgaps. However, an analysis of the
onset potentials of its p- and n-doping processes shows that
this reduction is mainly contributed by an increase in n-type
onset potential, although a minor decrease is also observed in
the onset potential of the p-doping process. This does not
agree very well with the prediction that the stronger elec-
tron-donating ability of the SR group should greatly raise the
energy level of the valence band, i.e., to lower the onset po-
tential of p-doping. We had attempted to find a possible rea-
son, for instance, the residue electron-withdrawing Br- or Cl-
substituent on the polymer backbone might be responsible for
the observation. However, the ratios of the aromatic rings to
Br or Cl atoms in the polymer chains of P1a and P2a are not
significantly lower than those of P1a0 and P2a0 (Table 2), thus
excluding this possibility.
30
the oxidation by FeCl3 (denoted as Srþ). The tailoring part
with its S(2p3=2) component possessing the highest BE at 166.7
eV is due to the oxidation of S atoms on the side chain into
sulfone groups.29
Electrochromism. The change in the UV-Vis-NIR spectra
during electrochemical oxidation or reduction provides direct
evidence for the doping processes, and gives information about
the electronic band structure of the polymers. Figure 4 shows
the UV-Vis-NIR spectra of P1a upon p-doping. When the
applied potential is higher than its onset value, the color of
the film changes from brown to black, and two new broad
absorption bands emerge at 630 nm and about 1600 nm. In-
creasing the potential causes an increase in the intensity of the
bands, together with a slight blue shift in the bandgap and a
Thus, the doublets with BE of S(2p3=2) at 163.1 and 166.7
eV are assigned to the neutral and sulfone species, which lo-
cate on the substituent group. However, their (Sa) sum con-
stituents about only 39% of the total sulfur (Table 6). This
suggests that some of the Sa may be involved in other envi-
ronments, in order to account for the theoretical 50% of the Sa
in total sulfur. A possible explanation had been put forward by
other researchers.10;31;32 Based on results obtained from con-
ductivity and absorption spectrum changes before and after
doping, they suggested the oxidation of pendant sulfur and
oxygen atoms into positively charged species by the dopants.
Our observation here can serve as direct evidence from the
viewpoint of XPS for the oxidation of sulfur atoms located
on the pendent chains.
ꢀ
slight decrease in the intensity of the p–p transition. These
features of optical transitions are due to the formation of
bipolarons.9{11
The n-dopability of the polymer films was also directly
shown by the spectral change measured in situ when cathodic

J. M. Xu et al.
Bull. Chem. Soc. Jpn., 76, No. 7 (2003) 1455
Fig. 6. XPS spectra of P2b: (a) S(2p) spectrum of neutral polymer; (b) S(2p) spectra of FeCl3-doped polymer; (c) Cl(2p) spectra of
FeCl3-doped polymer.
ethylenedithio-2,5-thiophenediyl).11 The I2-doped state is
Table 6. Summary of Results Relevant to Doping Level
thermally stable, because a thermal treatment of I2-doped P2
ꢃ
from XPS Analysis
at 100 C for 24 h showed no intensity decrease in the FTIR
Polymera) [Cl-Clc]4/[S]/%b) [Sa]/[S]/%c) [Saþ]/[Sa]/%d)
spectrum. However, the high doping level (I2 weight uptake
ꢅ 200 w/w% for P2 vs ꢅ 120 w/w% for P2018) and good
thermal stability did not make the P2 polymers more conduc-
P1a
P1b
P1c
P2a
P2b
P2c
8
9
5
47
48
49
39
39
29
6
4
2
22
22
42
tive, since their conductivities are mostly in the range of 10ꢁ4
10ꢁ5 S cmꢁ1 (Table 1).
–
29
30
49
In view of the high doping level indicated by the results
from elemental analyses, XPS and I2 weight uptake, the low
conductivities of the I2- and FeCl3-doped P2 series of poly-
mers are unexpected. The low conductivities together with the
high thermal stability of the doped states, especially when
doped by I2, seem to suggest that the bipolarons formed upon
doping are partially localized. This is in line with the XPS
result mentioned above, which suggests the presence of posi-
tively charged sulfur atoms on the pendant alkylthio side
chains.
a) Samples are doped by FeCl3. P2 are in powder form but P1
are films. b) The value of [Cl-Clc]4/[S] indicates the normal-
ized amount (in percentage) of charged chlorine species (Clꢁ)
corresponding to the total sulfur. Hence, 4 times of this value
affords the average number of dopants associated with one
monomeric repeating unit, as the latter contains 4 sulfur
atoms. c) Low value of [Sa]/[S] means that large percentage
of Sa are positively charged or the corresponding doping level
is high; on the contrary, high value indicates a low doping
level. This is reflected in the next column, the correponding
[Saþ]/[Sa]. d) [Saþ]/[Sa] = 2 ꢂ (50% ꢁ [Sa]/[S]). It is the
percentage of the positively charged Sa in the total Sa.
Conclusions
Two series of polymers comprising alkylthio (SR)-function-
alized thiophenediyl/bithiophenediyl and phenylene units had
been synthesized and characterized. They are highly fluores-
cent with emission located in the bluish green and green range
in solution and film states, respectively. The results, together
with those from literature reports, indicate that the insertion of
phenylene units is an efficient way to tune the emission color
and to increase the fluorescence quantum yield of conjugated
polymers. As compared to the alkyl(R)-functionalized ana-
logues, the SR-functionalized polymers have lower bandgap
energies through raising the reduction onset potentials and
comparative fluorescence quantum yields. This may imply
that replacing of R by SR did not bring higher coplanarity to
the polymer backbone, and the electronic effect of the SR
group was thus not effectively displayed. The doping levels
of the chemically oxidized SR-functionalized polymers are
high, as suggested by microanalysis and XPS results, but this
did not make them highly conductive. The oxidation of the
sulfur atoms on the SR side chains and, consequently, the
The Cl(2p) spectrum is presented in Fig. 6(c). As reported
earlier,20 the two doublets are attributed to the chlorine species
involved in the dopant (FeCl4ꢁ) and the C–Cl bond30 (denoted
as Clc), respectively. [Cl-Clc]4/[S] can also serve as a measure
of the doping level (Table 6). These results are in good agree-
ment with those obtained by elemental analysis (Table 2).20
P1 films have lower values of [Cl-Clc]4/[S] and [Saþ]/[Sa],
compared with those of P2. This is because the P1 samples
are in the film state, which can be dedoped easily.
Stability of the Doped States. The I2-doped P20 were
easily dedoped when the compressed pellets were ground into
powder.20 The dedoping resulted in the regeneration of un-
doped polymers, as indicated by TG and FTIR spectral data.
However, the I2-doped SR-substituted polymers are more
stable and the FTIR spectra (not shown) of I2-doped samples
show strong doping-induced features, as reported for poly(3,4-

1456 Bull. Chem. Soc. Jpn., 76, No. 7 (2003)
Alkylthio-functionalized Polymers
quartz cell comprising a film-coated ITO working-electrode, a
platinum counter electrode and a silver wire quasi-reference elec-
trode using the potentiostat together with a Lamda 900 UV-Vis-
NIR spectrometer in the same electrolyte system as CV.
localization of polarons and bipolarons may be responsible for
the observation.
Experimental
Materials.
Tetrahydrofuran (THF), dimethyl sulfoxide
Synthesis. The general schematic synthesis routes to the poly-
mers are shown in Scheme 1. The main monomers employed
were 1,4-bis(3-alkylthio-2-thienyl)benzene (4, alkyl = butyl, octyl
and dodecyl). While P1 polymers were prepared by polyconden-
sation of the Grignard reagents of the dibromination product of 4
with 1,4-dibromobenzene, P2 polymers were made by the chemi-
cal oxidation of 4 using FeCl3 as an oxidant.
(DMSO) and acetonitrile were carefully dried before use. 2-Bro-
mothiophene, 1,4-dibromobenzene, bis(diphenylphosphino)nickel
dichloride [Ni(dppp)Cl2], 1-butanethiol, 1-octanethiol and 1-do-
decanethiol were used as received. 1,4-Bis(2-thienyl)benzene (1)
was synthesized in a similar way as for 1,4-bis(3-alkyl-2-
thienyl)benzene20 using 2-bromothiophene and 1,4-dibromoben-
zene as starting materials.
1,4-Bis(3,5-dibromo-2-thienyl)benzene (2): 1 (2.42 g, 10.0
mmol) was dissolved in CHCl3–AcOH (30 mL, 2:1) in a round-
bottom flask, and bromine (3.32 g, 20.7 mmol) dissolved in the
mixture solvent was added dropwise. The second portion of bro-
mine (3.32 g, 20.7 mmol) was then added and stirred at room
tempeꢃrature for a few hours. The reaction mixture was refluxed
at 60 C overnight and then cooled to room temperature. It was
extracted with CHCl3 and washed by capacious water, sodium
carbonate, water, dried and the solvent removed. Recrystalliza-
tion from CHCl3–EtOH yielded a needle-form solid. Yield: 5.2 g
(94%); mp 201–202 ꢃC. MS (EI, m=e, % intensity): 554 (M, 64),
556 (M + 2, 95), 558 (M + 4, 100), 560 (M + 6, 96), 562 (M + 8,
68). HR-MS Calcd for C14H6S2Br4 (M + 4): 557.6603, Found
557.6629. IR (cmꢁ1): 3096, 3040, 3017, 1625, 1541, 1489, 1437,
1404, 1304, 837, 804, 611, 476. 1H NMR: ꢀ 7.65 (4H, s), 7.04
(2H, s). 13C NMR: ꢀ 138.94, 133.85, 132.13, 128.84, 112.08,
107.02. Anal. Calcd for C14H6S2Br4: C, 30.11; H, 1.07; S,
11.47; Br, 57.35%. Found: C, 29.84; H, 1.22; S, 11.29; Br,
57.35%.
Instrumentation. 1H NMR spectra were recorded on a 300
MHz Bruker ACF 300 FT-NMR spectrophotometer. 75 MHz
13C NMR spectra were obtained using the same instrument. Deu-
terated chloroform was used as a solvent and tetramethylsilane
(TMS) as an internal reference. EIMS and HRMS spectra were
obtained using a micromass 7034E Mass Spectrometer. Elemen-
tal analyses of all the monomers and polymers were conducted at
the NUS Microanalytical Laboratory on a Perkin-Elmer 240C ele-
mental analyzer for C, H, and S determination. Halogen determi-
nation was made by either ion chromatography or the oxygen flask
method. Gel permeation chromatography (GPC) analyses were
carried out using a Perkin Elmer Model 200 HPLC system with
PhenogelTM MXL and MXM columns (300 mm ꢂ 4.6 mm ID,
MW 100–100K and 5K–500K, respectively) calibrated using
polystyrene standards. THF was used as an eluent, and the flow
rate was 0.35 mL minꢁ1. The FTIR spectra of the polymers and
monomers dispersed in KBr disk was recorded on a Bio-Red TFS
156 spectrometer. Solution phase and solid state (films were
evaporatively cast from toluene solution) absorption and fluores-
cence spectrum measurements of the polymers were conducted on
a Hewlett Packard 8452A spectrophotometer and Shimadzu
RF5000 fluorescence spectrophotometer, respectively. Thermo-
gravimetry (TG) was conducted on a Du Pont Thermal Analyst
2100 system with a TGA 2950 thermogravimetric analyzer in air
1,4-Bis(3-bromo-2-thienyl)benzene (3): 2 (9.70 g, 17.4 mmol)
was mixed with a solution of 2-propanol (150 mL), AcOH (10
mL), water (2.8 mL) and concentrated HCl (0.6 mL). The mixture
was refluxed and stirred vigorously when zinc dust (5.48 g, 83.8
mmol) was added in small portions until the selective reduction of
2 to 3 was completed. TLC (hexane as the eluent) was employed
to monitor the progress of the reduction, by which Rf values of
0.64 were found for the starting material, 0.55 for the intermediate
tribromo-substituted compound and 0.45 for the product. The
reaction was completed and terminated by cooling to room tem-
perature about 8 h later, as indicated by the disappearance of the
first and the second points in TLC. The solvent was removed by a
rotavapor and the residue extracted with ether. The ether layer
was washed with water, dried, filtered and the solvent removed.
Recrystallization from ethanol afforded the titled product. Yield:
ꢃ
(75 mL minꢁ1) at a heating rate of 10 C minꢁ1 from room tem-
perature to 1000 ꢃC. Modulated differential scanning calorimetry
(MDSC) was conducted on a TA Instruments 2910 differential
scanning calorimeter upgraded with an MDSCTM option in N2
(70 mL minꢁ1) using a heating rate of 1 or 2 ꢃC minꢁ1, an oscilla-
tion amplitude of 1 or 2 ꢃC, respectively, and an oscillation period
of 60 or 80 s from ꢁ100 to 170 ꢃC unless otherwise stated.
Conductivity measurements were carried out on polymer pellets
of known thickness using a four-point probe connected to a Keith-
ley constant current source. XPS measurements of polymer pow-
der or films (on ITO) were performed by means of a VG ESCA/
SIMLAB MKII with a Mg Kꢃ radiation source (1253.6 eV). The
binding energies were corrected for surface charging by referenc-
ing to the designated hydrocarbon C(1s) binding energy as 284.6
eV.
Cyclic voltammetry of solution cast polymer films was carried
out using an EG&G 273A potentiostat/galvanostat controlled by
EG&G M270 research electrochemistry software under an argon
atmosphere at a scan rate of 50 mV sꢁ1. A three-electrode single-
compartment chemical cell consisting of an ITO glass plate as the
working-electrode, a platinum wire as the counter electrode, an
Ag/AgNO3 (0.1 M, in acetonitrile) (1 M = 1 mol dmꢁ3) as the
quasi-reference electrode [0.35 V versus a saturated calomel elec-
trode (SCE)] and a solution of 1 M Bu4NBF4 in acetonitrile as
electrolyte was employed. In situ electrochromism studies of
polymers were conducted in a single-compartment, three electrode
ꢃ
3.55 g (51%); mp 119–120 C. MS (EI, m=e, % intensity): 398
(M, 60), 400 (M + 2, 100), 402 (M + 4, 69). HR-MS Calcd for
C14H8S2Br2 (M + 2): 399.8414, Found 399.8389. IR (cmꢁ1):
3102, 3067, 3021, 1605, 1543, 1487, 1433, 1406, 1342, 870,
835, 696, 611. 1H NMR: ꢀ 7.73 (4H, s), 7.31 (2H, d, J ¼ 5:3
Hz), 7.07 (2H, d, J ¼ 5:3 Hz). 13C NMR: ꢀ 137.43, 132.59,
131.82, 128.89, 125.18, 107.73. Anal. Calcd for C14H8S2Br2:
C, 42.00; H, 2.00; S, 16.00; Br, 40.00%. Found: C, 41.63; H,
1.98; S, 16.31; Br, 40.28%.
General Procedure for the Preparation of 1,4-Bis(3-alkyl-
thio-2-thienyl)benzene (4). In a 250 mL two-neck flask, sodium
hydride (NaH, 60% dispersion in hexane, 0.96 g, 24 mmol) was
washed with dry hexane and blown with N2 to remove hexane.
DMSO (7 mL) was then added, followed by the addition of 1-
alkanethiol (24 mmol) dropwise. When the evolution of hydrogen
ceased, 3 (2.4 g, 6.0 mmol) and a mixture of CuO (0.096 g, 1.2
mmol) and KI (0.096 g, 0.58 mmol) in DMSO (4 mL) were

J. M. Xu et al.
Bull. Chem. Soc. Jpn., 76, No. 7 (2003) 1457
ꢃ
transferred. The mixture was refluxed at 120 C for 24 h under
nitrogen and then cooled down, filtered and washed with CHCl3.
The filtrate was extracted with CHCl3 and thoroughly washed
with water. Drying, filtering and removing CHCl3 led to a dark
oily residue. It was doubly isolated by silica-gel chromatography
using hexane–CHCl3 (5:1) as the eluent, and collected as the pro-
duct.
(4H, t, J ¼ 7:3 Hz), 1.53 (4H, m), 1.28 (20H, m), 0.86 (6H, t, J ¼
6:7 Hz). 13C NMR: ꢀ 141.61, 133.33, 132.46, 128.97, 128.79,
110.87, 35.69, 31.68, 29.25, 29.04, 28.97, 28.52, 22.54, 14.00.
Anal. Calcd for C30H40S4Br2: C, 52.33; H, 5.81; S, 18.60; Br,
23.26%. Found: C, 51.97; H, 5.87; S, 18.04; Br, 23.35%.
1,4-Bis(5-bromo-3-dodecylthio-2-thienyl)benzene (5c):
ꢃ
Yield: 82%; mp 64–65 C. MS (EI, m=e, % intensity): 798 (Mþ,
1,4-Bis(3-butylthio-2-thienyl)benzene (4a): Yield: 82%; mp
ꢃ
36), 800 (M + 2, 54), 802 (M + 4, 50), 43 (C3H7þ, 100). HR-MS
(peak match method) Calcd for C38H56S4Br2: 798.1631, Found
798.1626. IR (cmꢁ1): 3042, 3011, 2955, 2917, 2849, 1526,
1493, 1470, 1425, 1377, 870, 819, 717, 621, 461. 1H NMR: ꢀ
7.62 (4H, s), 7.04 (2H, s), 2.77 (4H, t, J ¼ 7:2 Hz), 1.54 (4H, m),
1.23 (36H, m), 0.87 (6H, t, J ¼ 6:6 Hz). 13C NMR: ꢀ 141.68,
133.37, 132.47, 128.96, 128.80, 110.85, 35.71, 31.81, 30.24,
29.54, 29.48, 29.25, 29.01, 28.50, 22.59, 14.02. Anal. Calcd for
C38H56S4Br2: C, 57.00; H, 7.00; S, 16.00; Br, 20.00%. Found: C,
56.76; H, 6.65; S, 15.74; Br, 19.43%.
42–43 C. MS (EI, m=e, % intensity): 418 (Mþ, 100). HR-MS
Calcd for C22H26S4: 418.0917, Found 418.0912. IR (cmꢁ1):
3098, 3044, 3013, 2955, 2926, 2868, 1527, 1470, 1460, 1431,
1377, 885, 831, 723, 630. 1H NMR: ꢀ 7.71 (4H, s), 7.29 (2H,
d, J ¼ 5:3 Hz), 7.10 (2H, d, J ¼ 5:2 Hz), 2.81 (4H, t, J ¼ 7:2 Hz),
1.53 (4H, m), 1.30 (4H, m), 0.85 (6H, t, J ¼ 7:3 Hz). 13C NMR: ꢀ
140.33, 133.03, 131.23, 128.86, 128.17, 124.02, 35.25, 31.41,
21.69, 13.51. Anal. Calcd for C22H26S4: C, 63.16; H, 6.22; S,
30.62%. Found: C, 62.86; H, 6.26; S, 30.15%.
1,4-Bis(3-octylthio-2-thienyl)benzene (4b): Yield: 74%; mp
þ
General Procedure for the Synthesis of Poly(3-alkylthio-2,5-
thiophenediyl-1,4-phenylene-alt-4-alkylthio-2,5-thiophenediyl-
1,4-phenylene) (P1). A solution of 5 (1.0 mmol) in THF (2 mL)
was introduced into a round-bottom flask containing Mg (0.060 g,
2.5 mmol) and THF (0.5 mL). The mixture was stirred vigorously
and refluxed under N2 for 2 h, while 1,2-dibromoethane (70 mL,
0.84 mmol) was added in portions as an initiator. The reaction
mixture was cooled down and a solution of 1,4-dibromobenzene
(0.236 g, 1.00 mmol) and Ni(dppp)Cl2 (10.8 mg, 0.0200 mmol) in
40–41 ꢃC. MS (EI, m=e, % intensity): 530 (Mþ, 38), 43 (C3H7
,
100). HR-MS Calcd for C30H42S4: 530.2169, Found 530.2178.
IR (cmꢁ1): 3096, 3042, 3013, 2955, 2924, 2870, 1526, 1483,
1466, 1433, 1377, 881, 837, 717, 627. 1H NMR: ꢀ 7.71 (4H, s),
7.29 (2H, d, J ¼ 5:2 Hz), 7.09 (2H, d, J ¼ 5:3 Hz), 2.80 (4H, t,
J ¼ 7:2 Hz), 1.53 (4H, m), 1.21 (20H, m), 0.85 (6H, t, J ¼ 6:8
Hz). 13C NMR: ꢀ 140.37, 133.05, 131.28, 128.85, 128.21, 123.96,
35.60, 31.66, 29.34, 29.01, 28.56, 22.50, 13.95. Anal. Calcd for
C30H42S4: C, 67.92; H, 7.92; S, 24.15%. Found: C, 67.59; H, 7.48;
S, 24.49%.
ꢃ
THF (2 mL) was then transferred at 0 C. The mixture was then
warmed up to room temperature and refluxed for 24 h before
being terminated by cooling down and pouring into methanol.
A brown sticky material was obtained which was subjected to
Soxhlet washing with acetone for 2 h.
1,4-Bis(3-dodecylthio-2-thienyl)benzene (4c): Yield: 75%;
ꢃ
mp 44–45 C. MS (EI, m=e, % intensity): 642 (Mþ, 100). HR-
MS Calcd for C38H58S4: 642.3421, Found 642.3415. IR (cmꢁ1):
3096, 3044, 3011, 2955, 2922, 2851, 1524, 1483, 1466, 1430,
1377, 883, 837, 719, 627. 1H NMR: ꢀ 7.71 (4H, s), 7.29 (2H,
d, J ¼ 5:2 Hz), 7.09 (2H, d, J ¼ 5:3 Hz), 2.80 (4H, t, J ¼ 7:2 Hz),
1.54 (4H, m), 1.23 (36H, m), 0.88 (6H, t, J ¼ 6:7 Hz). 13C NMR:
ꢀ 140.33, 133.04, 131.26, 128.86, 128.19, 123.99, 35.60, 31.82,
29.55, 29.50, 29.40, 29.35, 29.25, 29.07, 28.58, 22.60, 14.03.
General Procedure for the Preparation of 1,4-Bis(5-bromo-
3-alkylthio-2-thienyl)benzene (5). 4 (1.4 mmol) was dissolved
in CHCl3 (10 mL) in a two-neck round-bottom flask and bromine
(0.46 g, 2.9 mmol) in AcOH (60 mL) was run in dropwise. The
mixture was stirred at 40–50 ꢃC for 24 h, cooled down and dried.
After filtering and removing the solvent, a dark-brown oil was
obtained. It was isolated by flash chromatography (hexane as
the eluent) to offer a product that solidified at room temperature.
1,4-Bis(5-bromo-3-butylthio-2-thienyl)benzene (5a): Yield:
85%; mp 67–68 ꢃC. MS (EI, m=e, % intensity): 574 (M, 89), 576
(M + 2, 100), 578 (M + 4, 93). HR-MS Calcd for C22H24S4Br2
(M + 2): 575.9109, Found 575.9187. IR (cmꢁ1): 3040, 3013,
2953, 2922, 2866, 1529, 1483, 1462, 1419, 1369, 870, 819, 735,
625, 476. 1H NMR: ꢀ 7.62 (4H, s), 7.04 (2H, s), 2.78 (4H, t, J ¼
7:3 Hz), 1.52 (4H, m), 1.35 (4H, m), 0.85 (6H, t, J ¼ 7:2 Hz).
13C NMR: ꢀ 141.66, 133.36, 132.47, 128.98, 128.81, 110.87,
35.38, 31.33, 21.65, 13.48. Anal. Calcd for C22H24S4Br2: C,
45.83; H, 4.17; S, 22.22; Br, 27.78%. Found: C, 45.94; H, 4.61;
S, 22.33; Br, 27.28%.
Poly(3-butylthio-2,5-thiophenediyl-1,4-phenylene-alt-4-bu-
tylthio-2,5-thiophenediyl-1,4-phenylene) (P1a): Dark red pow-
der. Yield: 57%. 1H NMR: ꢀ 7.74 (4H, m), 7.64 (m), 7.30 (d, J ¼
5:2 Hz), 7.19 (1H, m), 7.10 (d, J ¼ 5:2 Hz), 2.82 (2H, m), 1.56
(2H, m), 1.37 (2H, m), 0.85 (3H, m). 13C NMR: ꢀ 134.77, 132.82,
131.25, 128.72, 127.60, 124.12, 35.32, 31.41, 21.72, 13.54. Anal.
Calcd for C14H14S2: C, 68.29; H, 5.69; S, 26.02; Br, 0%. Found:
C, 63.72; H, 6.11; S, 27.22; Br, 4.01%.
Poly(3-octylthio-2,5-thiophenediyl-1,4-phenylene-alt-4-oc-
tylthio-2,5-thiophenediyl-1,4-phenylene) (P1b): Red sticky ma-
1
terial. Yield: 50%. H NMR: ꢀ 7.74 (4H, m), 7.62 (m), 7.29 (m),
7.19 (1H, m), 7.10 (m), 7.03 (m), 2.83 (2H, m), 1.56 (2H, m), 1.23
(10H, m), 0.86 (3H, m). 13C NMR: ꢀ 134.84, 132.82, 131.30,
128.71, 127.66, 124.08, 35.68, 31.71, 29.34, 29.04, 28.60, 22.55,
14.00. Anal. Calcd for C18H22S2: C, 71.52; H, 7.28; S, 21.19; Br,
0%. Found: C, 62.21; H, 8.12; S, 19.91; Br, 6.43%.
Poly(3-dodecylthio-2,5-thiophenediyl-1,4-phenylene-alt-4-
dodecylthio-2,5-thiophenediyl-1,4-phenylene) (P1c): Red sticky
1
material. Yield: 10%. H NMR: ꢀ 7.73 (4H, m), 7.61 (m), 7.30
(m), 7.18 (1H, m), 7.10 (m), 2.85 (2H, m), 1.56 (2H, m), 1.23
(18H, m), 0.86 (3H, m).
General Procedure for the Synthesis of Poly[4,40-bis(alkyl-
thio)-2,20-bithiophene-5,50-diyl-alt-1,4-phenylene] (P2). They
were produced in the same way as P20 using FeCl3 as previously
reported.18
1,4-Bis(5-bromo-3-octylthio-2-thienyl)benzene (5b): Yield:
90%; mp 59–60 ꢃC. MS (EI, m=e, % intensity): 686 (Mþ, 34), 688
(M + 2, 63), 690 (M + 4, 50), 43 (C3H7þ, 100). HR-MS Calcd
for C30H40S4Br2 (M + 2): 688.0182, Found 688.0204. IR (cmꢁ1):
3038, 3015, 2955, 2928, 2851, 1524, 1491, 1460, 1423, 1377, 870,
819, 731, 623, 459. 1H NMR: ꢀ 7.62 (4H, s), 7.04 (2H, s), 2.77
Poly[4,40-bis(butylthio)-2,20-bithiophene-5,50-diyl-alt-1,4-
1
phenylene] (P2a): Yield: 83%. H NMR: ꢀ 7.74 (4H, m), 7.30
(d), 7.19 (2H, m), 7.10 (d), 2.86 (4H, m), 1.56 (4H, m), 1.38 (4H,
m), 0.88 (6H, m). Anal. Calcd for C11H12S2: C, 63.46; H, 5.77; S,
30.77%. Found: C, 62.65; H, 5.65; S, 33.05; Fe, 0.16; Cl, 0.75%.
Poly[4,40-bis(octylthio)-2,20-bithiophene-5,50-diyl-alt-1,4-

1458 Bull. Chem. Soc. Jpn., 76, No. 7 (2003)
Alkylthio-functionalized Polymers
1
phenylene] (P2b): Yield: 77%. H NMR: ꢀ 7.74 (4H, m), 7.30
12 J.-L. Bredas in ‘‘Handbook of Conducting Polymers,’’ 2nd
ed., ed by T. A. Skotheim, Marcel Dekker Inc., New York (1986).
13 R. H. Friend and N. C. Greenham in ‘‘Handbook of Con-
ducting Polymers,’’ 2nd ed., ed by T. A. Skotheim, R. L.
Elsenbaumer, and J. R. Reynolds, Marcel Dekker Inc., New York,
Basel, and Hong Kong (1998).
14 T. J. Kang, J. Y. Kim, K. J. Kim, C. Lee, and S. B. Rhee,
Synth. Met., 69, 377 (1995).
15 A. P. Davey, S. Elliott, O. O’Connor, and W. Blau, J.
Chem. Soc., Chem. Commun., 1995, 1433.
16 J. Li and Y. Pang, Macromolecules, 30, 7487 (1997).
17 S. C. Ng, T. T. Ong, and H. S. O. Chan, J. Mater. Chem., 8,
2663 (1998).
18 S. C. Ng, J. M. Xu, and H. S. O. Chan, Synth. Met., 92, 33
(1998).
(d, J ¼ 5:1 Hz), 7.18 (2H, m), 7.10 (d, J ¼ 5:2 Hz), 2.85 (4H, m),
1.56 (4H, m), 1.23 (20H, m), 0.86 (6H, m). Anal. Calcd for
C15H20S2: C, 68.18; H, 7.58; S, 24.24%. Found: C, 64.85; H,
9.02; S, 27.09; Fe, 0.23; Cl, 1.19%.
Poly[4,40-bis(dodecylthio)-2,20-bithiophene-5,50-diyl-alt-1,4-
phenylene] (P2c): This polymer was synthesized oxidatively as
P2c0, starting from 4c, but only a rubbery material was obtained,
even after 40 h. The rubbery crude product was dedoped, dried
and then oxidized a second time to offer a dark-green powdery
1
polymer. Yield: 56%. H NMR: ꢀ 7.74 (m), 7.19 (m), 7.09 (m),
2.84 (m), 1.55 (m), 1.25 (m), 0.86 (m). Anal. Calcd for C19H28S2:
C, 71.25; H, 8.75; S, 20.00%. Found: C, 64.27; H, 7.69; S, 23.44;
Fe, 0.13; Cl, 2.21%.
Doping of the Polymer. Polymers were doped by FeCl3 or I2
as reported earlier.18;20
19 S. C. Ng, J. M. Xu, H. S. O. Chan, A. Fujii, and K.
Yoshino, J. Mater. Chem., 9, 381 (1999).
20 S. C. Ng, J. M. Xu, and H. S. O. Chan, Macromolecules,
33, 7349 (2000).
21 A. Furlani, M. V. Russo, G. Polzonetti, K. Martin, H. H.
Wang, and J. R. Ferraro, Appl. Spectrosc., 44, 331 (1990).
22 M. V. Russo, G. Polzonetti, and A. Furlani, Synth. Met., 39,
291 (1991).
23 A. Pron, Z. Kucharski, C. Budrowski, M. Zagorska, S.
Krichene, J. Suwaiski, G. Dehe, and S. Lefrant, J. Chem. Phys.,
83, 5923 (1985).
Financial support from National University of Singapore
(NUS) under the research grant RP960613 is gratefully
acknowledged. JMX is grateful to the NUS for the award of
a research scholarship.
References
1 J. R. Elsenbaumer, K. Y. Jen, and R. Oobodi, Synth. Met.,
15, 169 (1986).
2
M. Sato, S. Tanaka, and K. Kaeriyama, J. Chem. Soc.,
Chem. Commun., 1986, 873.
24 T. J. Kang, J. Y. Kim, K. J. Kim, C. Lee, and S. B. Rhee,
Synth. Met., 69, 377 (1995).
3
(1993).
4
S.-A. Chen and C.-C. Tsai, Macromolecules, 26, 2234
25 M. Ueda, T. Hayakawa, O. Haba, H. Kawaguchi, and J.
Inoue, Macromolecules, 30, 7069 (1997).
26 D. R. Lide, ‘‘CRC Handbook of Chemistry and Physics,’’
CRC Press, Boca Raton, Boston, London, New York and
Washington D. C. (1998).
27 E. Block, ‘‘Reactions of Organosulfur Compounds,’’
Academic Press, New York, San Francisco and London (1982).
28 M. Mastragostino and L. Soddu, Electrochem. Acta, 35,
463 (1990).
29 J. F. Moulder, W. F. Stickle, P. E. Sobol, and K. D.
Bomben, ‘‘Handbook of X-ray Photoelectron Spectroscopy,’’ Per-
kin-Elmer, Eden Prairie (1992).
M. L. Bolhm, J. E. Pickett, and P. C. V. Dort, Macromol-
ecules, 26, 2704 (1993).
5 X. Wu, T.-A. Chen, and R. D. Rieke, Macromolecules, 29,
7671 (1996).
6
F. Goldoni, D. Iarossi, A. Mucci, L. Schenetti, and M.
Zambianch, J. Mater. Chem., 7, 593 (1997).
7
A. O. Patil, Y. Ikenoue, F. Wudl, and A. J. Heeger, J. Am.
Chem. Soc., 109, 1858 (1987).
D. J. Guerrero, X. Ren, and J. P. Ferraris, Chem. Mater., 6,
1437 (1994).
8
9 S. Tanaka, M. Sato, and K. Kaeriyama, Synth. Met., 25,
277 (1988).
30 E. T. Kang, K. G. Neoh, and K. L. Tan, Phys. Rew. B, 44,
10461 (1991).
10 J. P. Ruiz, K. Nayak, D. S. Marynick, and J. R. Reynolds,
Macromolecules, 22, 1231 (1989).
11 C. G. Wang, J. L. Schindler, C. R. Kannewurf, and M. G.
Kanatzidis, Chem. Mater., 7, 58 (1995).
31 K. Faid, M. Leclerc, M. Nguyen, and A. Diaz, Macromol-
ecules, 28, 284 (1995).
32 M. C. Gallazzi, L. Castellani, R. A. Marin, and G. Zerbi, J.
Polym. Sci., Part A: Polym. Chem., 31, 3339 (1993).